> I'm running HTTrack on a folder containing only data files 
> (for instance, <http://www.unl.edu/images/>), and though all 
> the files are mirrored just fine, I also get a bunch of 
> index files that are NOT in that directory. They're 
> strangely named, too: index5272.html, indexb3a7.html, and 
> so forth.

Yes, because you are also capturing the links at the top of the page (Name,
Last Modified, Size, Description) which dynamically present the directory
list.

Add this to your scan rules:
-*/?*