<p>Vehicular Adhoc Networks (VANETs) consist of vehicles as dynamic nodes and base stations, Wi-Fi access points as stationary nodes. These stationary nodes are also termed as Road Side Units (RSU). The communication occurs either among vehicles or between vehicles to roadside infrastructures or it may be hybrid too. Secured data communication is one of various challenging issues in VANET. Data trust and entity trust determine the reliability of data communication in the network. Always there is a possibility of malicious data communication if malicious and greedy nodes are present in the network. The network should be protected and preserved from different types of attacks and entities causing such attacks. So, there is a need to design trust models to improve the security in message communication in VANETs. The paper proposes a Hybrid Neuro Fuzzy Trust Management (NFTM) model which is a combined trust model for VANETs with distributed architecture. This model determines the Node Trust Value of nodes. Based on this NTV value, a node may be classified as trustworthy or not trust worthy. A trustworthy node is allowed to transmit the message whereas a non-trustworthy node is not allowed to do so. The node trust value may play a vital role in routing protocol. When a vehicle creates and sends a message, the node trust value is checked to find the trust worthiness of the node and that of the next nodes used for routing the packet to the destination. The firefly algorithm is used for optimized routing in which node trust value is considered as the glowing factor of the fireflies. The simulation results validate that the proposed NFTM model offers better solution for trust management and is cost-effective in terms of optimal routing as compared to the other trust models.</p>