The danger they face with "modularity" beyond that is how do they create products for people not using all of the modules?
It's tricky, but shouldn't be impossible. Basically, they should assume people are using the 'core' modules (whatever they are) but are not using any supplementary modules (except where absolutely essential). However, they should write in such a way that they don't
exclude people who are using whatever supplementary modules, again except where absolutely essential.
So if I'm playing with Modules, A, C and E, will an adventure product cover that? And lets say I'm playing an adventure that's released before Module E was available? Does that mean I have to do significant extra work as GM to make it work?
For the most part, I would expect published campaigns to use only the core modules, plus possibly a single module designed especially for that campaign (for example, the Birthright AP might include a domain management module). I don't expect there to be many exceptions to this - adventures are a hard sell at the best of times, so they'll need to make sure they exclude as few people as possible.
I would also expect to see the 'big' modules being produced first, with more niche modules coming later and having a smaller footprint in the game. That way, if using an adventure from before module E was published, although there will inevitably be some rework required, it will be kept to a minimum.
Of course, I also fully expect that most of the planned modules will never see the light of day, except perhaps in a very skeletal form in an "Unearthed Arcana" type of book.
Will the modularity significantly confuse people?
I suspect not - I would expect the core product to have few if any modules included, for size reasons if nothing else, and for those modules to be very 'light' in nature. For the more advanced players who are likely to seek out modules, it will probably be no different from the old model of "we're the Complete books and the Expanded Psionics Handbook, but absolutely not BoXD!".