One of the obvious problems with video-based instruction, especially the more visually captivating, ironically, is the potential for viewers to drop back into "TV-trance-mode", absorbing but not doing much processing or demonstrating meaningful engagement. (There is also a very serious issue with visual modality overpowering auditory and kinaesthetic, as well.) In pronunciation work, where re-education of the body is central, not enthusiastically joining "the dance" is a deal breaker . . . One key contribution of gesture to instruction is to create stronger engagement and enhancement of moment-by-moment attention.
A 2014 study, The effect of gestured instruction on the learning of physical causality problems by Carlson, Jacobs, Perry and Ruth Breckinridge-Church demonstrates how systematic use of gesture by instructors on video can significantly improve learning of another "physical" process. Subjects who viewed the "gesture-articulated" instructor, rather than just the spoken presentation did better on the post test. This study is particularly relevant in that it deals with gesture enabling cognition of what is a very "tactile" concept, that of manipulating gear movement and direction.
I came up with this system over a decade ago and still use videos (of myself) when introducing students to the basic gestural inventory, or pedagogical movement patterns (PMP). I'm just so much better online . . . (and you will be, too!)
Alibali, M., Young, A., Crooks, N., Yeo, A., Wolfgram, M., Ledesma, I., Nathan, M., Breckinridge Church, R. and E. Knuth. (2013). Students learn more when their teacher has learned to gesture effectively. Gesture 13:2, 210–233.
Carlson, C., Jacobs, S., Perry, M. and R. Breckinridge-Church. (2014). The effect of gestured instruction on the learning of physical causality problems. Gesture 14:1, 26–45.