The power train of recently introduced Range Extended Electric Vehicle (REEV) comprises an engine, two servo-controlled electric machines, three clutches, and a relatively large battery. In order to maximize the fuel efficiency and provide smooth driving, such a vehicle requires a sophisticated control strategy whose development and optimal tuning requires a widespread use of power train models of different complexity. The bond graph methodology is used in the paper to develop a unified mathematical model of the REEV power train dynamics. A special attention is devoted to the development of a computationally efficient clutch subsystem model. The model complexity can easily be adapted to different control system design tasks, ranging from low-level control and active damping strategies, through off-line optimization of control variables, to supervisory energy management control. The functionality of the model is illustrated on a comparative simulation example of electrical driving using one or two electrical machines.