In this paper, the focus is on modeling local risks and uncertainties by generalized spatial autoregressive conditional heteroscedasticity (spGARCH) models. In contrast to temporal ARCH models, in which the distribution is known given the full information set of the prior periods, the distribution is not straightforward in spatial and spatiotemporal settings. However, spatial GARCH models allow for instantaneous dependence in the local variance. Thus, the models are suitable to model spatial risk clusters. Furthermore, spatial GARCH models can be used to account for local differences in model uncertainties, if they are considered as error process of any spatial model, like spatial autoregressive or spatial regression models. As the first conditional and unconditional moments are zero, spGARCH models are flexible tools for modeling residuals without influencing the mean model. The particular aim of this paper is to analyze the effect implied by the above mentioned spatial GARCH-type models. Hence, we inspect the so-called spatial spill-over effects in the second moments via simulation studies. These spill-over effects describe how an increase in the local risk of one location spreads out across all regions nearby.