This study mainly focuses on the development of a self-organizing global sliding mode control for active power filters. First, a quasi global sliding mode control method is designed for inner current control loop to eliminate the reaching mode and chattering phenomenon. Then, the control law is constructed based on the meta-cognitive fuzzy neural network (MCFNN) rather than the actual systems to overcome the drawbacks of global sliding mode control. Different from the predefined structure approaches, only necessary data can be extracted to adjust the structure and parameters of the networks in the MCFNN. Subsequently, the Lyapunov stability analysis is presented to satisfy the control objectives and system stability requirements at the same time. Finally, simulation and experimental results demonstrate that the proposed control methods offer superior properties in both steady state and transient operation.