Mobile applications are one of the main features of mobile devices that have increased the prevalence of mobile use. In turn, such prevalence has increased the demand for high quality, interactive, and intelligent mobile applications. Achieving these requirements on multiple platforms, in critical domains, and with short time-to-market causes high challenges to developers and companies, which can be addressed by Model-Driven Development (MDD). The MDD approach plays a substantial role in increasing software productivity and enhancing solution quality by synthesising mobile applications from high-level specifications, and by automatically generating implementations on different platforms rather than relying on manually coding different platform versions. Effective application solutions strongly depend on the appropriate choice of architecture design and back-end services, and expertise in these choices can be encoded into an MDD process. Although some current research attempts to adopt MDD for mobile applications, the studies are still limited and have room for improvement. The published work tends to focus on either generating user interfaces or on simple data-centric applications. Therefore, this research aims to support the solution quality of complex cross-platform mobile applications by developing an advanced MDD framework that is compatible with native mobile applications, supports complex algorithms, as well as advanced services such as machine learning and cloud-based mobile back-ends. The inputs required for this framework are textual models for the user interface, data, and process of the app, and the outputs are complete functional mobile applications for Android and iOS.