Most existing techniques to handle strict task priorities in hierarchical control are based on null-space projectors or a sequence of quadratic programs; whereas non strict task priorities are usually handled by optimization based on a weighting strategy. This paper proposes a novel approach to handle both strict and non-strict priorities of an arbitrary number of tasks, and to achieve multiple priority rearrangements simultaneously. A generalized projector, which makes it possible to completely project a task into the null-space of a set of tasks, while partially projecting it into the null-space of some other tasks, is developed for priority modulation. The control input is computed by solving one quadratic programming problem, where generalized projectors are adopted to maintain a task hierarchy, and constraints can be implemented. The effectiveness of this approach is demonstrated on a simulated robotic manipulator in a dynamic environment.