As a Ph.D. candidate working in parsing and interested in model-theoretic syntax, I have found this book to be very useful. I have become very interested in the relation between languages, automata and logic, and how they relate to parsing and deduction. This handbook presents these things with some extra tidbits in chapters which from what I have read - I haven't read the whole book - are well-written and accessible. And the authors point out connections between each other's work.
What is especially interesting to me is the focus on languages where the chosen models are not strings, but instead trees, graphs, etc. For example, my present work is about semiring parsing, with tree automata as the operational model. I am interested in applying parsing as deduction to models which are arbitrary classes of graphs described by grammars written as formulas in some logic. Having recently bought this handbook for future reference, I was delighted to find a very accessible and useful chapter on tree automata which related to work I just started.
For my interests, the chapters on graph grammars, term rewriting and on automata, languages and logic are also enticing and I look forward to reading them as well. The other chapters aren't central to my own work, but I also look forward to reading them, for they seem interesting.
I encourage any researcher working with languages, whether they be artificial ones like XML trees, graphs representing networks or proofs, or they are natural languages, who would like to enter the new millenium to read this book as well as the other volumes of the Handbook of Formal Languages.