This study explores the electrical properties of 3D-printed carbon nanotube (CNT)-cementitious nanocomposites using a dual modelling approach that integrates micromechanics with finite element (FE) analysis for self-sensing applications. 3D-printed cementitious structures are prone to early crack formation, and incorporating CNTs enhances the material’s electrical conductivity, enabling a built-in health monitoring system. Given the critical impact of CNT dispersion, waviness, and orientation on the electrical conductivity, a theoretical model was developed to incorporate these factors. Experimental tests were conducted to evaluate the material’s conductivity and mechanical performance, and to validate the model. The results show that CNTs align with the printing direction at a 3D-printed layer’s surface, while maintaining isotropic behaviour at the core. Additionally, although the CNTs aggregate, they do not entangle with each other, instead creating conductive networks within the bundles. Moreover, the CNTs are not straight but wavy, which affects the material’s electrical conductivity. Based on the experimental results and the analytical model, a finite element model was developed to predict the conductivity of printed layers, accounting for varying CNT orientations throughout the layer. The results indicate that the model overestimates 3D-printed materials’ conductivity, suggesting that further studies are needed to account for interlayer effects on the measurements.