The Modular Multilevel Converter-Battery Energy Storage System (MMC-BESS) is a converter system with the potential to enhance grid reliability. However, the implementation of model predictive control (MPC) in MMC-BESS is confronted with challenges such as substantial computational demands, suboptimal quality of output current, and challenges in selecting weighting factors. To address these challenges, this paper puts forward an extended-level MPC approach, which involves the direct calculation of the output voltage reference value in the load MPC stage, followed by the optimization of the output level through a step-by-step optimization method. Additionally, it employs a cost function that optimizes the current area to further reduce the total harmonic distortion (THD). Notably, this process ensures a computational complexity of 5 for each control cycle, independent of the number of sub-modules. Two control strategies are employed to optimize the circulating current control depending on the number of inserted sub-modules. The use of independent cost functions for output current and circulating current eliminates the need for weighting factors, while the circulating current feedback for balancing the state-of-charge (SOC) is incorporated into the circulating current reference to achieve inter-phase and bridge arm SOC balancing control. The effectiveness of the proposed method is validated by simulation and experimental results.