Traffic congestion in cities creates a substantial challenge that results in operational delays and raises fuel consumption and creates environmental issues. Existing traffic management systems demonstrate poor performance when adapting to current time conditions which causes higher system inefficiencies. The research examines how new traffic management solutions should utilize deep learning and reinforcement learning together with fuzzy logic to achieve transformation. This document shows how to implement the collective power of AI models to process traffic information with optimized signal controls while reacting to traffic pattern changes. SUMO-based simulation demonstrates how the suggested approach functions effectively. The AI-based approach proves effective by decreasing waiting periods and improving road traffic movement at a higher level than traditional solutions. This study demonstrates how Artificial Intelligence enables the creation of adaptive and sustainable traffic control systems with smart capabilities. Pursuing these next steps involves integrating Artificial Intelligence with self-driving vehicles in addition to applying this technology across all smart urban applications.