Two teaching methodologies are presented and compared in this study: on the one hand, semi-guided tours in immersive virtual reality and, on the other, viewing video renderings of 3D environments. The two techniques are contrasted through 3D modeling of a fifteenth-century Spanish town called Briviesca, in an immersive environment, viewed with Oculus Rift. The suitability of virtual reality for teaching is assessed through questions on historical knowledge and urban layout. The understanding of the undergraduate students is evaluated, through questionnaires, after the viewing sessions. The responses of the students underline the effectiveness of the two methodologies: Video screenings received higher scores for historical ideas and the virtual tour was the most effective method at conveying knowledge learnt while viewing. Additionally, two user movements for controlling the virtual reality environment were tested: (1) gamepad locomotion and (2) roomscale movements combined with teleporting. The clear advantage of the second option was the total lack of motion sickness effects. However, the natural tendency using teleporting was to move very quickly through the city areas with no singular buildings and to spend more time in front of these types of buildings. They therefore missed visual information related to the first areas while retaining more information related to those buildings. Finally, the spatial location of singular buildings was clearly better acquired with the virtual tour.