There are several reasons I prefer to use the velocity addition formula. Basically, it addresses the problem in terms of kinetmatics - how we describe motion, independent of forces and masses. It illustrates that the problem of reaching c is kinematical in nature, and not dynamical. People who are confused about the issue often think they can get around the limit of the speed of light by reducing the mass of the object somehow. This won't work, and the velocity addition explanation explains why.
The additional concepts needed for dynamics (masses and forces) can be introduced at a later date. The dynamical explanation for the speed of light limit is superfically attractive at first glance, but leads to later confusion - see any of the threads about "relativistic mass" vs "invariant mass" for the sorts of confusion generated.
[add]
Basically, it's better to treat dynamics properly, than to give an incorrect half-baked introduction to relativistic dyanamics to students too early, one that basically has to be "unlearned" because it was not properly built in the first place. The problem of why c is the limiting velocity does not have to use any of the concepts of dynamics at all - the concepts of kinematics are sufficient.