IN DEFENSE OF TENNOKOE BANK
I dunno, man, I don't think there's anything too complicated about the individual-file solution, especially with a little basic idiot-proofing of the interface. In fact, I think it's easier in a lot of ways.
Let's take a look at the problem the Tennokoe Bank causes. Imagine you have Save-A, Save-B and Save-C in your internal memory. You want to keep Save-A and Save-B on deck, but you don't need Save-C at the moment; you just don't want to delete it.
Now along comes Save-D, from a new game you want to play, and it won't fit into the internal memory. If the Tennokoe Bank worked like I think it should, you would be able to copy out Save-C to make room, and that would be that. Save-A, Save-B and Save-D would all be happily in the internal memory for you to have easy access to, and Save-C would be quickly and efficiently tucked away.
Instead, however, the reality is that you have to put Save-A, Save-B and Save-C
all into the Tennokoe, and now the only thing in your internal memory is Save-D. Want to play the game that uses Save-A, then a little more of Save-D? That's two more memory swaps.
And then there's the temptation to do something really stupid: copy Save-A/B/C to the Tennokoe Bank without actually swapping, and then just delete Save-C from the internal memory to make room for Save-D. Now you've got two copies of A and B floating around, only one of which you're going to make progress with. Heaven help you if you get them mixed up a couple of weeks later, and especially if you then go on to delete something without checking it.
It's worth mentioning that since there are four banks, that's four times as much potentially wasted space on the Tennokoe since you probably can't fill each bank perfectly.
It's all this versus just copying single files back and forth from a single big bank. There's no complex file system or anything, either, it's just two lists: internal memory and Tennokoe memory.
Sounds like a piece of cake to me.