The advocacy of task-based language teaching (TBLT) has met with resistance. The critiques of TBLT and the misconceptions that underlie them have already been addressed in Ellis (2009) and Long (2016). The purpose of this article is to move forward by examining a number of real problems that TBLT faces--such as how a "task" should be defined, the nature and timing of the "focus-on-form" required, how to determine task complexity and sequence tasks, the role of explicit instruction, the timing of focus on form and the teacher training needed to ensure effective uptake of TBLT--and to also consider what solutions are possible. Disagreements exist regarding the relative merits of task-based and task-supported language teaching. I will propose that a hybrid/modular syllabus that allows for a balance between an object-oriented and a tool-oriented view of language teaching offers the most promising way forward.