Reepithelialization of skin wounds is essential to restore barrier function and prevent infection. This process requires coordination of keratinocyte proliferation, migration, and differentiation, which may be impeded by various extrinsic and host-dependent factors. Deep, full-thickness wounds, e.g., burns, are often grafted with dermal matrices before transplantation of split-skin grafts. These dermal matrices need to be integrated in the host skin and serve as a substrate for neoepidermis formation. Systematic preclinical analysis of keratinocyte migration on established and experimental matrices has been hampered by the lack of suitable in vitro model systems. Here, we developed an in vitro full-thickness wound healing model in tissue-engineered human skin that allowed analysis of the reepithelialization process across different grafted dermal substitutes. We observed strong differences between porous and nonporous matrices, the latter being superior for reepithelialization. This finding was corroborated in rodent wound healing models. The model was optimized using lentivirus-transduced keratinocytes expressing enhanced green fluorescent protein and by the addition of human blood, which accelerated keratinocyte migration underneath the clot. Our model shows great potential for preclinical evaluation of tissue-engineered dermal substitutes in a medium-throughput format, thereby obviating the use of large numbers of experimental animals.