I'm going to split this into two, for the 2nd and 3rd movements respectively. Also, I'm going to address the composition stuff first, and then the technical stuff.
--- 2nd mov ---
For the composition itself:
1) I feel sometimes that there's just too much attempt to keep "things going on," throughout the entire movement. Specially the constant 8th/etc figures that go through the piece are good for effect, but you abuse them a little too much when you really didn't need them in a lot of sections. Remember, less is sometimes more. By having less crap going on you can accentuate specific harmonies better.
Good example is measure 40 to 50. It would've been better to get rid of the bass figure at this point specially considering how you've been hammering the same model for a while now. The harmony you have there is good, but it shifts into background at this point since the most present thing is the bass movement. Don't throw the entire army on the listener like that all at once, remember the piano can only do so much in terms of dynamic variance and having already a FF go to FFF and be an audible difference means that FF must be already softer, and so on.
2) Which brings me to your dynamics. I would've pulled back from some of the "climatic" moments you did, specially considering how in many sections you have sudden contrasts and shifts (measure 53.) This means something like measure 43 for example not being F, but instead maybe sfz or subito piano. Otherwise, again like the point above, you hammer on the same dynamic for too long.
3) You're throwing so many ideas at the listener that really it's hard to hear the relation between the parts. In some instances it's a great idea and it's very well worked out, (72 to the arpeggio model in 86.) But sometimes the change doesn't work as well as it could've because you're using elements you tired beforehand (the 72 repetition model doesn't work as well because you already had the "ongoing notes" figure going on for the whole section beforehand.)
But even so, a lot of the sections are linked really well and it doesn't become really that much of an issue listening to it.
4) I like the overall form of the slow movt, being mostly a single crescendo/written out accel through the entire movement. It works. However, I would insert more contrasts specially in the end (106 onwards) specially because here you also lose the characteristic harmony/writing you had at the beginning, which I don't think is necessary even if you're writing fast passages. But also, there's the issue of measure 50 being too bombastic too early in the piece and the piece never really recovering from it. It almost feels like you wrote yourself into a corner there in that everything that comes after measure 50 is almost a long drawn out ending. Maybe it would've been better to push that moment further back, or to take some weight off it.
5) Related to the above, there's a certain lack of consequence in the harmonic choices you pick at the beginning. I'll point out a couple of instances that should have more consequence SPECIALLY in the ending stretch, where your writing turns out rather flat. Measures 19 until 29 are all very very full of dissonances and chromatic, which DO show up again later if only in passing, but never again with such central position (due in part to your piano writing, but more on that later.)
Another instance that springs to mind is precisely the shift in measure 70 till 72. It's a very strong harmonic passage, made even stronger by the change in dynamic, but it sticks out because of that. You could've worked it in near the end again in some form or another. In some ways, passages like those that have strong harmonic choices that create a lot of contrast draw much more attention to themselves than your "climax" passages (say, measure 106, 50, etc) and they also provide critical fresh air to your writing. Don't be afraid to use them more often.
6) There's a LOT of dissonance going on in the entire movement, and even in some areas you experiment with different chord anatomies that come I suppose as a result of voice leading (Measures 21, 51, etc) but I think that you could do a little more balancing. There's a point where adding a 6th and a 7th+9th to EVERY chord starts to become dull. Mix up your options, there are suspensions, passing notes, etc that you can do to very simple chords to get specially complex sound out of them without just jamming in the characteristic dissonances all the time. This is much more nuanced in the beginning of the movement. Sometimes less is more.
The portion of 50 till 56 is particularly guilty of this. And, also, if you'll be using so much dissonance, you can also spread it out across the piano rather than keep it all on the right hand. Remember that the way you distribute the voices changes how dissonance work (a 9th is never the same as a 2nd, etc.)
---
To the technical:
1) Dynamics. Dynamics dynamics dynamics. Stuff like what happens in measure 35 to 41 is nonsense. You have two crescendos to F, but no indication as to where the pianist has to cut off so that the second crescendo actually is audible. Write in as MUCH detail as POSSIBLE, where EVERY SINGLE dynamic goes.
Besides from that, and this is something I usually tell everyone, remember that if you have F, FF and FFF, F must be THAT much softer so that FFF can be two levels louder. In other words, this pushes F into almost mf range if you want the FFF to be audible as the peak. Visualize your softest and then your loudest dynamic and see if you can cut out as much "buffer" dynamics from in between. Is that FF section really that much louder than F? Is the FFF section really that much louder than the FF section? If so, just how loud is it? Bang on the piano loud?
Remember also that the louder the dynamic the harder it is to play quickly.
But then look please at measure 105. See that hopeless hairpin from FF to FFF? I know what your IDEA is, but do you realize how pointless it is in practice? How much is there to "crescendo" to when you're ALREADY banging on the piano in FF? Just like I was saying before, less is more. Keep in mind that writing a softer dynamic doesn't necessarily mean the actual sound will be soft (imagine that!) If I write a piece that mostly remains in ppp, then I write a single p, that p has to be that much louder than ppp. Get it? Same here. Limit yourself writing dynamic that is both audible and practical.
The entire section you have as FFF I would've simply left it at FF but added a > mark, or perhaps sfz. It is enough to tell the pianist "Ok this has to be loud at this point" but giving the entire SECTION a FFF dynamic is not such a good idea because it means the entire FF section has to be that much softer.
In this way your hairpins are also more meaningful, becuase the pianist isn't going to have to calculate how much he has to decresc before the hairpin is audible. In measure 105 the only way anyone would realistically play that is by shifting the dynamic at the beginning of the hairpin to F or maybe even softer, otherwise you won't hear it.
2) Take a look at measure 69. Second half, right hand. How fast are those notes, again?
Seriously, there's a lot of segments like those where, honestly, you could simplify it. In fact, your denser writing seems wholly out of place compared to the beginning, specially since measure 69 isn't that far away from where the piano was playing almost single chords! Go over your parts and think through just WHAT you need in terms of notes, and take the other stuff out.
3) Why pedal markings on 86?? And then the weird pedal in 93? Honestly no good pianist is going to pedal it any differently than what you wrote save for measure 93. In fact, the moment he pedals in 93 the sustained bass notes will go away making a bizarre hole in the bass. Is this what you intended? Otherwise just take out the pedal marks altogether, they don't tell anyone anything new or special. The only reason I could see for that pedal in 93 is because you're taking into account using the bass sustain and the general sustain... but honestly now, again, that scraggy is trivial for any good pianist. Don't need to write it out.
--- 3rd mov ---
Ok so now to the minuet's composition.
1) The entire thing is more like themes and variations than a proper "minuet," but even then it's still pretty much a bow form with a reprise at the end. The pacing is pretty alright, it's more balanced out than the 2nd.
2) It's a much less dissonant-heavy movement compared to the one before, and in here the sections where the harmony is interesting or different stand out in a good way rather than feeling underused. But at the same time, I think you held back a little too much though given the nature of what a minuet sonata ending is like it's alright.
3) The general pause works great. Maybe you could've used more throughout the piece?
---
Technical:
1) The same stuff I said before applies here but to a lesser extent. The piano writing is smaller, if more conservative. It works better overall though it suffers a little from the "fast notes, must fill up with stuff!" syndrome but less.
2) Otherwise, the less is more works good here. Your lines are simpler and it works better for it. This is alright.
---
Overall, I expect the first movement to be just as good if not better, good work.