The use of carbon fibre-reinforced composites has been growing exponentially in the past few decades. They offer excellent mechanical properties in combination with a low density, making them an ideal solution for many lightweight applications. However, they often suffer from a lack of toughness. In contrast with carbon fibre composites, self-reinforced composites have an excellent toughness, but a relatively low stiffness and strength. They consist of a polymer fibre in a matrix made from the same polymer. This thesis aims to break through the typical stiffness-toughness dilemma by hybridising carbon fibres with self-reinforced polypropylene (SRPP) and to design a material that is both stiff and tough. The focus lies on optimising the tensile properties and impact resistance of these novel hybrid composites.Before hybridising SRPP, it is vital that the influence of the process parameters on the mechanical properties of SRPP is understood. Hot compaction uses oriented monocomponent polymer tapes, and melts their outer surface to create the matrix. This process has a narrow processing window and is therefore inherently sensitive to the process parameters. It was shown that increasing the temperature or dwell time increased the matrix fraction and molecular relaxation of the oriented polymer tapes. This leads to improved interlayer bonding, which has a small effect on the tensile properties, but a large effect on the impact resistance. The compaction pressure was even more important, as too low of a pressure can strongly reduce the penetration impact resistance. The impact resistance is a key advantage of SRPP, and therefore potential issues with traditional testing techniques were identified. Hybridisation of SRPP with carbon fibres resulted in a novel class of hybrid composites with a unique combination of stiffness, strength, ultimate failure strain and impact resistance. Inter- and intralayer hybrids were developed and optimised. For interlayer hybrids, it was revealed how the damage development in tension can be controlled by changing the carbon fibre volume fraction, the carbon fibre and SRPP orientation, and the relative layer thickness. An appropriate choice of these parameters leads to pseudo-ductility, where the carbon fibre layers are able to fracture multiple times. For intralayer hybrids, the importance of intralayer bonding was highlighted. This parameter is crucial as a strong intralayer bonding reduces the ultimate failure strain and impact resistance, but improves the flexural properties. Improving the adhesion between carbon fibre and polypropylene has a similar, but more pronounced effect.The experimental work was supported by extensive modelling studies. A novel and versatile strength model for unidirectional hybrid composites was developed. This model was first elaborated for non-hybrid composites, and its strengths and limitations were identified. An in-depth experimental validation was performed for carbon fibre composites by comparing fibre failure predictions with synchrotron computed tomography data. This led to vital recommendations for future model developments. This model was then extended to hybrid composites and an extensive parametric study was performed. This study focused on the hybrid effect, which is a synergistic effect that increases the failure strain of carbon fibres through hybridisation with a more ductile fibre. The hybrid effect was shown to increase by reducing the carbon fibre volume fraction and by improving the dispersion of both fibre types. The mechanical properties of the ductile fibre were not crucial for the hybrid effect, provided its failure strain is at least twice as high as the carbon fibre failure strain. Using very ductile polypropylene fibres instead of the traditional glass fibres hence does not lead to a larger hybrid effect in carbon fibre hybrid composites. The main advantage of polypropylene fibres is its potential of achieving a larger ultimate failure strain. The predictions of the hybrid effect were also compared to experimental measurements. This validation was the first of its kind to achieve a good agreement, which indicates that the model captures the main phenomena of the hybrid effect.Finally, a road map for optimising hybrid self-reinforced composites was set up by combining the experimental results with the modelling insights. This road map can also be used to optimise other hybrid self-reinforced composites. The presented results revealed the potential benefits of hybrid composites. They should provide a driving force for future work on hybrid composites and for improvements in processing technologies for manufacturing well-dispersed hybrid composites.