Well, I thought we were talking about such checks dictating how PCs behave… the major thrust of your argument is about how you know and understand the character better than any mechanical process could replicate. So I don’t think what you’re describing really addresses that in any way.
As for the Persuasion check example, I would expect that the GM would perhaps factor several of those things into theDC of the check, and whether or not to apply advantage or disadvantage to the roll. If a GM literally approached the DC and ad/disad the same way for all 30,000 different permutations of the conversation, I’d say that’s poor GMing. Or, perhaps, the consequence of such a limited system.
I don’t think your concern expressed here even really applies to D&D, let alone other RPGs.