Breast tissue consists of an epithelial parenchyma embedded in stroma, of heterogeneous and complex composition, undergoing several morphological and functional alterations throughout females' lifespan. Improved knowledge on the crosstalk between parenchymal and stromal mammary cells should provide important insights on breast tissue dynamics, both under healthy and diseased states. Here, we describe an advanced 3D model of breast tissue, combining multiple components, namely stromal cells and their extracellular matrix (ECM), as well as parenchymal epithelial cells, in a hybrid system. To build the model, porous scaffolds were produced by extrusion 3D printing of peptide-modified alginate hydrogels, and then populated with human mammary fibroblasts. Seeded fibroblasts were able to adhere, spread and produce endogenous ECM, providing adequate coverage of the scaffold surface, without obstructing the pores. On a second stage, a peptide-modified alginate pre-gel laden with mammary gland epithelial cells was used to fill the scaffold's pores, forming a hydrogel by ionic crosslinking. Throughout time, epithelial cells formed prototypical mammary acini-like structures, in close proximity with fibroblasts and their ECM. This generated a heterotypic 3D model that partially recreates both stromal and parenchymal compartments of breast tissue, promoting cell-cell and cell-matrix crosstalk. Furthermore, the hybrid system could be easily dissolved for cell recovery and subsequent analysis by standard cellular/molecular assays. In particular, we show that retrieved cell populations could be discriminated by flow cytometry using cell-type specific markers. This integrative 3D model stands out as a promising platform for studying breast stroma-parenchyma interactions, both under physiological and pathological settings.