The integration of 2D visualization and navigation techniques has reached a state where the potential for improvements is relatively low. With 3D-stereoscopy-compatible technology now commonplace not only in research but also in many households, the need for better 3D visualization and navigation techniques has increased. Nevertheless, for the representation of many abstract data such as networks, 2D visualization remains the primary choice. But often such abstract data is associated with spatial data, thereby increasing the need for combining both 2D and 3D visualization and navigation techniques. Here, we discuss a new hybrid-dimensional approach integrating 2D and 3D (stereoscopic) visualization as well as navigation into a semi-immersive virtual environment. This approach is compared to classical 6DOF navigation techniques. Three scientific as well as educational applications are presented: an educational car model, a plant simulation data exploration, and a cellular model with network exploration, each of these combining spatial with associated abstract data.