Using fuzzy logic, we propose a model with a neural network for public transport, normal cars, and motorcycles. The model controls traffic-light systems to reduce traffic congestion and help vehicles with high priority pass through. A fuzzy neural network (FNN) calculates the traffic-light system and extends or terminates the green signal according to the traffic situation at the given junction while also computing from adjacent intersections. In the presence of public transports, the system decides which signal(s) should be red and how much of an extension should be given to green signals for the priority-based vehicle. The system also monitors the density of car flows and makes real-time decisions accordingly. In order to verify the proposed design algorithm, we adapted the simulations of sumo, ns2, and GLD to our model, and further results depict the performance of the proposed FNN in handling traffic congestion and priority-based traffic. The promising results present the efficiency and the scope of the proposed multi-module architecture for future development in traffic control.