I know a lot about PAF 5, but I certainly don't know everything.
What evidence do you have that multiple parents aren't reported accurately after 10 generations? (And is this the case in the database itself or in one particular report generated from the database?)
I am unfamiliar with the term "sosa N°". I can't find it listed in PAF's Help file. A search on Google failed to find a definition. But Google did find a few French websites using the term, although none that I looked at defined the term. Asking Google to translate "Numero sosa" results in "Sosa Number" ... again no definition is given for sosa.