The demands for high data rate, reliability, high energy efficiency, high spectral efficiency, and low latency communication have been increasing rapidly. For this reason, communication models that use limited resources in the best way, allow fast data transmission, and increase performance has become very important. In this work, a novel high energy and spectral efficient reconfigurable intelligent surface aided spatial media-based modulation system, called RIS-SMBM, is proposed for Rayleigh fading channels. In addition to the bits carried in the M-QAM symbol, while media-based modulation (MBM) provides data bits to be carried in the indices of different channels according to the radio frequency (RF) mirrors are on or off, spatial modulation (SM) provides data bits to be carried in the indices of the transmit antennas. By combining these two modulation schemes, the spectral efficiency increases considerably since the amount of information transmitted in the same time interval is substantially increased. The optimal maximum-likelihood (ML) detector and the enhanced low-complexity (ELC) detector for the RIS-SMBM system are proposed. The ELC detector achieves near ML performance while reducing the complexity of the optimal ML detector for the proposed RIS-SMBM system. We analyze the average bit error rate (ABER), throughput, complexity, and energy efficiency for the RIS-SMBM scheme and verify the analytical results with Monte Carlo simulations. It has been observed that the proposed system provides better error performance as well as providing higher spectral and energy efficiency than benchmark systems.