Hybrid nanofluids contain more than one type of nanoparticle and have shown improved thermofluidic properties compared to more conventional ones that contain a single nanocomponent. Such hybrid systems have been introduced to improve further the thermal and mass transport properties of nanoparticulate systems that affect a multitude of applications. The impact of a second particle type on the effective thermal conductivity of nanofluids is investigated here using the reconstruction of particle configurations and prediction of thermal efficiency with meshless methods, placing emphasis on the role of particle aggregation. An algorithm to obtain particle clusters of the core–shell type is presented as an alternative to random mixing. The method offers rapid, controlled reconstruction of clustered systems with tailored properties, such as the fractal dimension, the average number of particles per aggregate, and the distribution of distinct particle types within the aggregates. The nanoparticle dispersion conditions are found to have a major impact on the thermal properties of hybrid nanofluids. Specifically, the spatial distribution of the two particle types within the aggregates and the shape of the aggregates, as described by their fractal dimension, are shown to affect strongly the conductivity of the nanofluid even at low volume fractions. Cluster configurations made up of a high-conducting core and a low-conducting shell were found to be advantageous for conduction. Low fractal dimension aggregates favored the creation of long continuous pathways across the nanofluid and increased conductivity.