To my knowledge Hofstadter never actually talks about multiplicity (though he does come quite a bit closer than this in some of his other books). Daniel Dennett (the other author of The Mind's I) does talk about multiplicity elsewhere, in Speaking for Ourselves (http://pp.kpnet.fi/seirioa/cdenn/speaking.htm) and also in the 13th chapter of Consciousness Explained. The chapter in Consciousness Explained is probably worth reading: it's a lot better than the earlier Speaking, and to my mind it's one of the few acceptable philosophical considerations of multiplicity from a materialist viewpoint (i.e. not involving souls).

Cognitive scientists in general are probably going to have the view of the mind as recursively decomposable from modules into submodules into subsubmodules, because the field is greatly influenced by computer science and that is how computer programs look. This is even more true of artificial intelligence researchers like Hofstadter.