In this paper, a new hybrid framework for 3D scene modeling is proposed, which combines the efficiency of NeRFs with the reproducibility of explicit techniques. Despite the fact that NeRF has produced quite an impressive result in synthesizing scenes from 2D images, in complex scenes the method has its drawbacks in handling spatial dimensions and occlusions. To address these issues, our proposed Unified NeRF approach adds explicit dimensions to steer the neural radiance field, improving performance across different scenes and rendering qualities. This hybrid approach has been thoroughly evaluated for performance on a number of benchmark datasets. The outcome seen was that this approach outperforms the traditional NeRF models especially in scenes with complex geometries or dynamic parts. This research responds to some of the most pressing challenges in 3D scene modeling and opens gates to further investigation of the integrated approach of both explicit and implicit methods.