That's a pretty insightful point.
Maybe a more sensible way would be that when you "send" a fleet, you're really just queueing it and it remains active in your main fleet until the next tick starts. So there's no reason to wait until the very last minute anymore, *and* all fleets queued on a given tick all depart at the same time (start of next tick) and arrive at the same time.
Of course, that would mean every fleet gains an extra tick in transit, as the first tick of travel no longer happens during the same time as the player's action. It would no longer be possible to attack on the next tick, you'd always have at least 1 full tick to travel to non-portalled destinations. To "fix" that, we could simply negate the first tick of travel for every fleet sent, and attribute it to "hyperdrive acceleration" or something similar that fits the game's narrative.
Regardless, having queued fleets remain in your active fleet for a given tick would have a pretty severe impact on how people play the game currently, but I don't think that's necessarily a bad thing.
It sounds like we're touching on multiple points here, which is great. I do think it's worthwhile to question things that "just are" because IC has been the same since forever.
Going back to the non-instant fleet return specifically, this would definitely slow down certain aspects of the game but it would also give more reason to manage your fleets effectively. If we combined this with something like the possibility to split/merge fleets and re-route fleets during transit, we have an entirely new game mechanic.
Of course, that's biting off quite a bit. At the very least, we could try simply requiring fleets to travel back to the nearest portal, and make it galaxy setting so if we don't like it we can just disable it.
Lots to think about!
Got a few bucks? The Imperial Tip Jar is accepting contributions!