It absolutely makes sense, depending on your perspective: If it's the amount/combinations of civs you can have interact with at any given time within a match or when starting a match, then yes, the raw quantity of civs per era matter. The combinations of leaders and civs isn't relevant because (as far as I know), you can't have a match with duplicate civs but different leaders. And even if you can, that doesn't really matter as an option within certain contexts, such as for players who don't want to do that or only want somewhat-fitting leader choices per civs (which tbh isn't really even possible right now at all, which is a problem)
Fundamentally I would not consider civs being per-era a good tradeoff, even just from the perspective of # of civs and ignoring the gameplay implications, how it kinda screws over Prehispanic civilizations as not having later era reps (which is a huge deal to me) etc, unless the game ends up having at least 30 playable civs per era. I realize that is a LOT of content to develop for, especially when civs in civ 7 have more unique bonuses, units/buildings etc then in the past, and when you'll only be able to access that content for 1/3 (or 1/4th if there's an atomic era) of the game, but 30 civs per era/at any one time is still notably less then what Civ 5 or especially what Civ 6 ended up having once all the DLC and expansions are done.
If even doing that much is too much effort, and I get it very well might be, then I have to question, again, if the tradeoff of having civ switching is worth the effort: It seems like a big increase in development workload, with it arguably giving players less content (at least as much as you could argue it gives us more), and also creates a lot of fundamental historical/thematic and roleplaying issues for players.