Sometimes it's by asking questions that you get a better overall picture. After examinig the second half of the first measure of Bach's Brandeburghese concerto n²5, here is what i've found,
beat num: 1 2 3 4 5 6 7 8 6) violino: C# D C# B A G F# E 5) viola: E D E C# 1) bass: A B C# A 0) fig: 6
A standard do-ré-mi-do progression on A major dominant in the key of D major.
On beat number 1, A major
On beat number 5, A6 as figured.
What is interesting is what happens between the chords, in particular on beat 3 - B D C# - a 93 triad resolving to 83 on the next beat - B D B.
This is not what is expected here, canonically it should be a 643, a little sixth, so I suppose we're more in a Fuxian type of approach at this level of linearity.
So problem solved, I should read the Fux more closely, no espcape. Unless there is a school of counterpoint sctrictly based on figured bass, but I have a doubt about it. There is the diminution paradigm, but how does it apply to multiple lines? Fux again, I guess.