High speed synchronous reluctance machines can be used in electric drives and power generation systems. This paper develops nonlinear mathematical models for synchronous reluctance motors using the machine (abc) variables as well as the quadrature, direct and zero (qd0) variables. Kirchhoff's and Newton's laws are used to derive the differential equations that describe the circuitry and torsional-mechanical dynamics. The Park transformation is applied in the model developments to derive the nonlinear differential equations in the rotor reference frame. As complete, highly coupled, nonlinear mathematical models are found, synchronous reluctance motors should be controlled to attain the desired performances in the full torque-speed operating envelope. Hence, controllers must be designed. A robust controller with nonlinear error and state feedback mappings is designed to guarantee robust tracking and to ensure stability, as well as to attain disturbance rejection. The reported straightforward and computationally efficient synthesis procedure is based on the well understood nonlinear feedback mappings and Lyapunov stability theory. Innovative results in modeling, simulation analysis and design are applied to control electric drives driven by three phase synchronous reluctance motors.