White matter (WM) tracts shape the brain’s dynamical activity, and their damage (e.g., disconnections) results in functional alterations and cognitive symptoms [1]. Here, we use neural mass modeling to test the in-vivo effects of WM disconnections [2]. We create models of local versus global, and edge versus nodal effects of WM disconnections in a large fMRI dataset comprising individuals affected by varying degrees of white matter hyperintensities (WMH), in order to estimate the impact of WMH on brain dynamics in a data-driven manner. We use a virtual-lesioning (disconnectome [3]) approach informed by empirical data to account for these heterogeneous effects either on WM tracts (i.e., edges) or on dynamics (i.e. nodes). We compare these hypotheses to those arising from biological findings linking WMH to a global – yet invisible with conventional MRI – damage to WM tracts [4]. We show that, although WMH mainly diminish the structural strength of local interregional communication, they yield relevant global dynamical effects by reducing the overall synchronization of the whole-brain network. The effects on regional node dynamics are less relevant and evident only at later stages. This study provides evidence that WMH might be the tip-of-the iceberg of a widespread damage to WM tracts, also affecting the normal appearing white matter, with relevant global consequences on interregional communication. Furthermore, our study supports the relevance of whole-brain modeling for hypothesis testing of structure-function relationships and for integrating empirical data for model parameter estimation, bridging methodological advances from computational neuroscience and neuroimaging.