Ack, I should've known better.
I stayed up a bit later on Monday night than I intended to. I was asked, by Claude, last week, about whether certain cases (in particular the Born-Infeld model) not covered by my Skyrmion/wave-map pre-print can be done in the same way. At that time I wasn't thinking clearly, so the answer I gave was rather off-target. On Monday afternoon, the question crept back into my mind so I started obsessing over it. On Monday evening I had a faint inkling of idea how the whole thing can be solved, and so I stayed up late to write it down on my notepad.
Tuesday morning, after definitely not enough sleep (even though I slept in a bit until 9:30), I copied those notes down and re-worked them into the Skyrmion paper and replaced it on arXiv. And then I started looking at other things. Last night, as I fell asleep, I kept getting this nagging feeling that the way I wrote some stuff was possibly not the best.
So this morning, on the way in to work, I realized how stupid I was. I made a lot of extraneous assumptions that, while at first glance seemed necessary, on hindsight is completely frivolous. I guess the moral of the story is: before judging whether a project is completed to satisfaction, make sure that my head is clear enough to make the call.