Navigation and selection are the two interaction tasks often needed for the manipulation of an object in a synthetic world. An interface that supports automatic navigation and selection may increase the realism of a virtual reality (VR) system. Such an engrossing interface of a VR system is possible by incorporating machine learning (ML) into the realm of the virtual environment (VE). The use of intelligence in VR systems, however, is a milestone yet to be achieved to make seamless realism in a VE possible. To improve the believability of an intelligent virtual agent (IVA), this research work presents DOP (Discover Objects and Paths), a novel model for automated navigation and selection. The model, by intermingling ML with the VE, intends to augment the maturity of a virtual agent to the extent of human-level intelligence. Using ML classifiers, an IVA learns objects of interest along with the paths leading to the objects. To access any known object, the IVA then follows a mental map of the scene for self-directed navigation. After reaching a proper location in the designed VE, the required object is selected by using the ML algorithms. Extending ML to VR, the model was implemented in a case-study project called Learn Objects on a Path (LOOP). The application, having a maze-like VE, was evaluated in terms of accuracy and applicability by eight users. The results obtained showed that the model can be incorporated into a number of cross-modality applications.