...or even somehow in Greek where the "to be" lacks the aorist/imperfect distinction. (as the only(??*) verb)
A great question, and, as you suspect, there are no distinctions! The Czech kids learning Latin are usually taught a quick "hack" that whatever in Czech is the perfect [lexical] aspect, gets rendered by the Latin perfect, otherwise by imperfect and vice versa and therefore people start to be clueless and quite "random" anytime they try to understand why Latin would use or even require some habuit or fuit in certain places.
Now, of course (and I could write a long comparative essay about this in here, I suppose ) there is a great great great problem in comparing the slavonic lexical aspect to the IE perfect(aorist)/imperfect distinction (they share some 'core', but... sometimes also drastically differ in some important aspects). So, the hack I described is very prone to error generation and by no way it substitutes the true conscious understanding of how the IE perfect(aorist) and imperfect really work.
A great question, anyway!
*I'm not so good in Greek, so I don't know if it's the only one
Thank you Godmy. This surprised me as Spanish and French do make the distinction (fue / era, elle a été / elle était), so I was puzzled when I saw, in another forum, a Czech speaker who is learning in Spanish asking questions about these tenses from a rather clueless base. If Czech doesn't distinguish aspect in stative verbs like these, then it all makes sense.