To investigate light propagation and near-field effects above structured surfaces, scanning near-field optical microscopy is a powerful tool providing access to the near-field intensity. These measurements can be combined with rigorous solving of Maxwell's equations to gain insight into light propagation inside the sample, which is not accessible via experiment. However, we find differences between the intensity distribution obtained via experiment and that observed in the simulation at a constant distance of 20 nm above the surface, which corresponds to the typical surface-to-probe distance in the experiment. A first explanation was given by topographic artefacts [Proc. SPIE 8789, 87890I (2013)]. To better understand the interaction between sample and probe in regard to light propagation, we include the probe in high-resolution simulations of different structures, with the position of the (finite-sized) probe resulting from its placement above each structure. While there is a visible difference in the overall light distribution of the system, caused by the probe, the relative intensity at the position of the probe is shown to be in very good agreement to the intensity in a system without the probe. This has been found for many probe positions along the surface of the structure. This result is applicable to many systems in different fields of research which use such measurements for obtaining information about near-field effects of samples. We show an application for thin-film photovoltaics, where light scattering textured surfaces are used to increase the path length of photons in the absorber layer to increase device performance.