We consider a multi-user (MU) full-duplex (FD) multiple-input multiple-output (MIMO) communication system, in which the base station transceiver is equipped with transmit and receive position reconfigurable antennas (PRAs) to mitigate both MU-interference (MUI) and self-interference (SI) while enhancing the desired signal quality. We first formulate a joint design of beamforming, transmit power allocation, and antenna placement to maximize the weighted sum-rate of the considered PRA-based MU-FD-MIMO system under the constraints on reconfigurable region size, minimum antenna spacing, and transmit power. To address this highly non-convex problem, we then propose an alternating optimization (AO) framework that decomposes the original problem into subproblems and solves them iteratively. In particular, fractional programming techniques are used to separate the optimization variables from the logarithmic and ratio terms, while the block successive upper bound minimization method handles the non-convexity of PRA placement. Simulation results confirm a significant performance gain when integrating PRAs into the MU-FD-MIMO system and demonstrate the advantage of the proposed optimization framework.