Windows are a complex part of building design and provide a considerable benefit, including to school buildings. For the evaluation of the daylighting conditions prevailing in classrooms, the daylight factor (DF) was considered as the most appropriate parameter for indicating the quantity of admitted daylight. The DF values and CIE overcast sky were calculated using Velux Daylight Visualizer 3 software. The task of the paper is to compare various roof window openings in relation to the level of daylight in the attic, looking to optimize the use of the attic for teaching. The indoor air temperature has a general influence on comfort in the interior, in addition to daylight. In winter, the situation is not critical. The thermal insulation properties of packaging structures are sufficient. The situation is worse in summer, due to the fact that the heat-storage properties are undersized and there is excessive overheating of the indoor air. Four variants of roof windows and their influence on the overall microclimate in the attic are compared. The variant without roof windows is a suitable solution with regard to minimum overheating, but the worst situation for daylight. In order to receive even more light from the window (by moving windows to the top of the roof), we can use variant 2. Based on a combination of daylight calculations and summer temperature, a graphical dependence on window size prediction in terms of top and combined lighting is derived. This was hypothesized without shading the windows. Of course, the shading elements of these windows or cooling are expected in the summer. Finally, the energy required for cooling is compared depending on the size of the windows and achievement of the permissible temperature.