Well to be fair this game was made for a 72 hour game jam. With that in mind I'm going to choose polish over expansion, especially the closer to the end of the time limit we get.
Most of your rational seems to be based off having the time to fully plan the scope and game all at once and have everything settled before making it. While for a game jam (especially such a short one) our initial build plan is to just make minimum viable. Then if you have the time, expand.
The main downside to this is that any major system that is added after a minimum viable build has to be built on top of already implemented code. Due to that even a small scope expansion can take many hours of re-writing code to get the new system to work on top of the old. The upside is that if something doesn't work or you run out of time you have the ability to backpedal to a working system.
The examples I gave for add-ons to the game are just possibilities, which would greatly change the game and thusly also change how we would describe it. Right now it is arcade-like, because that is the closest way we want to describe the game. If we made it so you would buy upgrades after each loss, we would then probably describe it as roguelike.
If we were to expand the game sometime in the future, we would probably change many, many things about it. There is a good possibility that we would start a brand new build that can be built with a much more defined idea in mind. That way all the systems can work together well, we would also have the time to test out additions and see how they feel before making them final. Then at the end finding what's best to describe the game.
Again, thank you for taking the time to give feedback on the game. What you have said and gone over will useful for future games we make, especially when deciding on an initial scope.