Modularization in XHTML 1.1 (Why)

I'm a beginning student in XHTML (1.1) now, but I have significant
experience in other, high-level programming languages, such as C
and C++. 

I would like to hear from PROGRAMMERS about the concept of
modularization in XHTML 1.1 where a designer/programmer includes
or excludes certain of the functional modules, based on the
functionality of the program/product that is being developed. 

Simply put, with C++ for example, you could obtain certain
functionality for your program by linking to code libraries, but the
addition of each library increased the size of your code. 

Thinking, as I do, that these XHTML modules are nothing but DTD
"references", that is, there is no actual code or overhead added to
your program, what does it matter if a development author qualifies
which modules of XHTML his product supports? 

What is gained by editing a "standard" DTD in order to define the
functionality that a given program or product supports, that is, if a
hand-held widget did not make use of FRAMES functionality, what is
the negative effect if its referenced DTD includes the FRAMES module
definition? 

I like the concept of modules and highly-defined structures, but how
does the practice of modularization actually affect the programmer?


I really would appreciate any information that the w3-savvy
programmers have to say about this issue. 

Thanks,