The opening of windows can lead to high energy losses in wintertime, especially in nearly zero-energy buildings. But can reduce overheating significantly in summertime. Therefore, window use models have been created in the past to assess the energy use and thermal comfort in residential buildings. The models are mostly based on weather-variables. However, a recent study (Verbruggen, Janssens, et al. 2018) indicated that these models were not able to accurately predict the window use in wintertime. For that reason, an occupancy and activity based model was developed. In this article, the impact of the application of the new window opening model on the residential energy use and thermal comfort was assessed. The object-oriented modelling language Modelica was used to simulate the energy use and temperatures in a nearly-zero energy house, which is a representation of an existing house in a nearly zero-energy neighbourhood in Kortrijk. From this neighbourhood, measured energy use data was available as well as window sensor data for some of the houses. These measured data were compared to the simulated data of the new window use model, a weather-based model and the Belgian EPBD-calculation method. The occupancy and activity based model could predict more accurately the average opening durations in wintertime and could better account for the large variation in window use compared to weather-based models. An optimal window opening strategy could limit the overheating significantly, even prevent it in the bedrooms and bathroom. However, opening the windows also implies an increase in energy use for heating. Some combinations of different window opening habits can limit the overheating, while limiting the increase in energy use at the same time.