This paper addresses the problem of self-triggered state feedback control for linear plants under bounded disturbances. In a self-triggered scenario, the controller is allowed to choose when the next sampling time should occur and does so based on the current sampled state and on a priori knowledge about the plant. The main contribution of this paper is the definition of a triggering strategy that allows for the consideration of a class of controllers that is much larger than that of static controllers with a zero-order hold of the last state measurement. This is done by resorting to a model-based control architecture whereby a model of the plant is used to simulate the time evolution of the plant’s state in between sampling times when no information about the plant’s actual state is available. An illustrative example with simulation results shows how the new triggering strategy along with a model-based controller design yield a clear gain in performance when compared to other strategies proposed so far.