Pneumatic Artificial Muscles (PAMs) have emerged as promising actuators in robotics and biomechanical applications due to their high power-to-weight ratio, compliance, and lightweight properties. However, accurately modeling and controlling PAMs remains challenging due to their inherent nonlinearity, hysteresis, and time-dependent dynamics. To address these challenges, this study presents a dynamic Takagi-Sugeno fuzzy model that leverages the Nonlinear Auto Regressive Moving Average with eXogenous input framework for improved system identification. The proposed model effectively captures the complex behavior of opposing PAMs without relying on predefined mathematical formulations. Experimental results demonstrate that the model achieves low prediction errors, making it well-suited for practical implementations.