This paper presents a binocular camera alignment system and a neural model integrated in the system. The neural model is based on the occipital lobe of the mammalian cerebral cortex responsible for primary visual input processing. The final goal is to achieve a brain-like integrated cognitive system. Although the components of such a system may have weaker performance than existing industrial solutions, the integrated cognitive system has a performance that has not yet been passed. The proposed neural model, as a component of a cognitive system, has a complex structure compared to solutions using engineering operators and algorithms. This complexity is due to the use of a vast number of simple and uniform neuronal computational units. In contrast with complexity, the possibility of parallel computation becomes a promising advantage. The proposed model takes the idea of the binocular, disparity selective cells in the visual cortex. The model is able to identify the disparity between an image pair, and decide in which direction the image sensors should be aligned to achieve binocular convergence. The model has a similar structure and functional characteristics as the network of binocular cells in the visual cortex. A PC-based implementation and integration of the model into a camera system has shown that the performed camera alignment is similar to the binocular convergence of biological systems.