This paper presents a general approach to modeling VBR (variable bit rate) compressed full-motion video. The salient feature of such video is the existence of scenes. Scene structure presents special problems if one is to construct high-fidelity source models to drive Monte Carlo simulations. Intuitively, scene transitions are often accompanied by a substantial change of bit rate (the size of compressed pictures), as well as the level of burstiness. This paper argues that a modulated model can handle well scene modeling, and illustrates this point by an example.