Glad it worked.

To try to answer your points/questions...

Probably the best resource for finding answers, or even pointers to answers,
is to search these newsgroups. Personally I prefer to do it via
http://groups.google.com. Most of the stuff I know about mailmerge and
fields has been posted in this group, for example, in some cases, many, many
times.

Books oriented towards "end users" typically cover whatever the authors
perceive are the most-used features in a product, except for books which
specifically cover "gotchas." I assume that a product like Word is not
generally regarded as "data-oriented" and so Mailmerge is typically reduced
to an explanation of what it is and some simple walkthroughs. I think that
is a pity because there are som many "gotchas" in mailmerge that plenty of
users will not be satisfied by the out-of-the-box features. Also, writing
about features that rely on interoperability is inherently difficult,
because both the author and the reader typically have to understand a fair
amount about other application involved, and the number of version-related
issues potentially skyrockets. There aren't many "technical" books about
Word that cover this area in depth -  in fact the only two that spring to
mind are both in German and authored by Cindy Meister, myself, and others.
Have a look at http://www.amazon.de if you are interested!

The vast majority of my knowledge, such as it is, comes from trial and error
on my part. Some of it is from reading contributions by other people,
especially when someone undermines a long-held belief.

The main criteria for learning stuff by trial and error in my experience are
a. be willing
b. have the necessary tools (e.g. you can't really deduce stuff about Word
97 without having a copy to hand)
c. know what you are trying to solve or discover
d. be wary of "it can't be done" assertions unless they come from someone
you trust to get that right
e. don't let preconceptions override the evidence of your own eyes (it is
rare, but not unknown, for people in this group to reject a suggestion
because they "can't see how it could work" or some such. My approach is "see
if it works, even if it seems weird, then try to work out why."
f. be tenacious, but also aware of the Law of Diminishing Returns.