Recently, room-temperature magnetic refrigeration (MR) has attracted immense attention as a prospective substitute for vapor-compression refrigeration. The greatest obstacles faced by existing room-temperature MR prototypes are their inadequate cooling capacity, small temperature span, and low efficiency. A potential solution to these issues could include constructing new MR cycles based on regeneration. Therefore, in this paper, we constructed a few potentially efficient two-stage MR thermodynamic cycles. The thermodynamic model of the two-stage MR cycle was established to evaluate its cyclic performance. The performances of different two-stage MR cycles were investigated, and compared with that of the basic MR cycle. The results revealed that the two-stage MR cycles always performed better than the basic MR cycle, on account of the increased cooling capacity, and the reduced magnetic work. Moreover, it was found that the two-stage MR cycle with two-stage magnetization and demagnetization was the optimal configuration. Therefore, further investigation was conducted on the performance of the optimal two-stage MR cycle. The effects of certain key parameters (such as, the intermediate magnetic field, and the cold- and hot-end operating temperatures) on the cyclic performance were carefully analyzed. It was observed that when the operating temperature was between 280 K and 310 K, a magnetic field ratio between 0.371 and 0.492 was optimal for achieving the best cyclic performance. Maximum thermodynamic perfection of approximately 0.635 was obtained, which was 27.55% greater than that of the basic MR cycle. Furthermore, we investigated the influence of adiabatic irreversibility on the cyclic performance. The results demonstrated that when adiabatic irreversibility factors ranged between 0.9 and 1, the two-stage MR cycle consistently performed better than the basic MR cycle. Thus, this paper can provide a theoretical direction for the design and optimization of efficient room-temperature magnetic refrigerators employing new thermodynamic cycles.