Efficient broadcasting in Mobile Ad Hoc Networks (MANETs) is challenged by dynamic topology. Traditional fuzzy logic-based schemes use static fuzzy tables and membership functions so they have limited their adaptability to changing network conditions. To solve this problem, We propose a novel broadcasting scheme that combines dynamic fuzzy logic membership functions with Reinforcement Learning (RL) techniques, specifically Temporal Difference (TD) learning and Monte Carlo (MC) methods. We dynamically adjust the membership functions based on network parameters like topology density, connection time of the nodes, and signal strength. The Reinforcement Learning component refines the fuzzy rule base by simulating network operations and updating the table using a reward and punishment mechanism. Simulation results demonstrate that our scheme significantly reduces redundant broadcasts and collisions, achieving lower network overhead and improved message delivery reliability in a dynamic environment compared to traditional static methods.