Wireless energy harvesting sensor networks constitute a new paradigm, where the motes deployed in the field are no longer constrained by the limited battery resource, but are able to re-charge themselves through directed electromagnetic energy transfer. The energy sources, which we call actors, are mobile and move along pre-decided patterns while radiating an appropriate level of energy, sufficient enough to charge the sensors at an acceptable rate. This is the first work that investigates the impact of energy transfer, especially concerning the energy gain in the sensors, the energy spent by the actors, and the overall lifetime in the resulting mobile sensor-actor networks. We propose two event-specific mobility models, where the events occur at the centers of a Voronoi tessellation, and the actors move along either (i)the edges of the Voronoi cells, or (ii) directly from one event center to another. We undertake a comprehensive simulation based study using traces obtained from our experimental energy harvesting circuits powering Mica2 motes. Our results reveal several non-intuitive outcomes, and provide guidelines on which mobility model may be adopted based on the distribution of the events and actors.