Well, it has certainly far from solved the problem, either way. Now that I have a kernel source tree to compile from, though, I have a few debugging tricks that I'll want to try and see if I can extract some information from.
In case anyone was wondering, by the way, I've discovered that the source of the lag has changes slightly from before the original OS upgrade, though. It's still a matter of ridiculously long I/O latency on simple operations, but whereas it used to stop in functions like close(), open(), rename() and the like, the by far most common source of lag right now seems to be doing mmap'ed reads from BSDDB region files.