This letter tackles spacecraft optimal control problems in which the cost function is defined by a sum of vector norms, in order to optimize fuel consumption while achieving sparse actuation. A model predictive control (MPC) strategy is devised for such type of problems, accounting for different spacecraft maneuvering modes. Closed-loop stability is guaranteed by a conic Lyapunov function, which is employed as a terminal cost in the formulation. A systematic method to construct such function is presented. The proposed design is compared to a standard quadratic MPC scheme on a long-range rendez-vous mission.