Manned-unmanned teaming (MUM-T) is still the most practical way to carry out real-world tasks due to the interdependence between humans and machines. This paper provides insights into developing modern MUM-T with the development of artificial intelligence (AI) and human-machine interaction (HMI) technologies. First, we discuss the current methods from the perspective of the OODA loop (Observe, Orient, Decide, Act), which is essential for understanding the task situation, making correct decisions, and responding as planned. We note that the human-in-the-loop manner is still necessary for making responsible decisions. Besides, intelligent interfaces for MUM-T are also crucial for ensuring that unmanned systems accurately execute human commands through natural user interfaces such as augmented reality (AR), speech recognition and dialogue systems, eye-tracking, and even for the task of human-swarm interaction (HSI). Finally, we discuss the design elements for MUM-T, including adjusting the levels of autonomy, optimizing the HMI interfaces, strengthening trust and cooperation, real-time feedback, and learning.