In the light of sustained urbanization and rising temperature trends, mitigating the impact of extreme heat events is a pressing urban planning issue. Radiation heat load, quantified as mean radiant temperature (Tmrt), has been identified as the main source of summer heat stress. Several studies ascertained that Tmrt is the key factor driving human thermal comfort in outdoor urban places. Shading that reduces radiation heat load (Tmrt) is the most effective means to mitigate heat stress in outdoor public places, which offer a venue for leisure, recreation and for the social life of residents. Nonetheless, the small-scale thermal conditions of urban places are not only governed by shade trees and greenery, but also by buildings and paved surfaces. The aim of this study is twofold. First, it assesses the impact of differently oriented street facades with varying solar exposure on the human radiation balance at a medium-sized square in a mid-latitude city. Second, it evaluates the performance of popular, freely available microclimate and radiation models in deriving Tmrt values. The well-vegetated, rectangular Bartók square was selected in Szeged, Hungary to calculate various radiation parameters using three numerical simulation models (ENVI-met, SOLWEIG and RayMan). The derived parameters (radiation flux densities from different directions, as well as Tmrt) are compared with corresponding values obtained from detailed on-site measurements. The field data are collected as part of a 24-hour long radiation measurement utilizing the six-directional method, where a set of pyranometers and pyrgeometers are used to record short- and long-wave flux densities from six perpendicular directions (from above, from below as well as from the four cardinal points). The model-measurement comparison is based on hourly data from five locations within the square: from the center and from near the four bordering street facades of the rectangular square. Our initial results indicate that besides direct solar radiation, the temperatures of artificial surfaces (e.i. of building walls and pavements) strongly influences the human radiation balance: the increased temperature of surrounding surfaces increases the amount of emitted long-wave radiation and thus, reduces the amount a person is able to dissipate. Investigations like ours are necessary both for the advancement of our filed in general, and for the development of numerical models in particular. Models are simplifications of reality and thus they introduce a certain degree of idealization: trees are never as perfectly shaped or have a homogeneous crown transmissivity and leaf area index (LAI) in the reality, neither do surface parameters are as uniform as frequently assumed by models. All these differences influence model results to a certain degree. Therefore, the ideal outdoor thermal conditions that practitioners often plan for from behind their desks are likely to be worse in reality. Identifying the strengths and weaknesses of different models and revealing how they compare to reality is essential for both scientists and urban planners, since they all need to understand and acknowledge the limitations of the numerical approach.