Ok. A slight re-think and a re-draw.
I've kept the image the same size for the captures and just moved the parts of the 3D model. I also included a tape measure so you see the range of movement this design has, but it looks like the tape didn't pick up on the OPEN image. Provided the laces are flexible like polyprop or similar, you should have plenty of room to get a foot in or out of the shoe. This system only needs 3 magnets points -
1. On the open end of each lace.
2. As a stop point when the laces are OPEN (about 95mm) so they don't come all the way out.
3. As a stop point when the laces are CLOSED (about 30mm).
So long as the laces slide easy, this would be a fast open/close system with no electronics and minimal moving parts.
http://www.therpf.com/attachment.php?attachmentid=680905&d=1478482791
http://www.therpf.com/attachment.php?attachmentid=680906&d=1478482791
Sorry, I have had monthly meetings that have kept me away from playing with this. I like the structure you are building. The path of the laces appears straight across when viewed from the top, and the actual retraction path is angled backwards. Part of my redesign was to provide plenty of clearance inside the tracks to accommodate this "trajectory shift" (probably not the best description). I see in your later post where you are advocating for the rounding of the edges, and I think that is savvy. If you are going to eliminate the elastic bands, then I can see how they would definitely benefit from the rounded edges.
I had another method I played around in my mind with. And truth be told, it is in process as a viable method in alternate to my other design. I have a friend who is ate up in the brain stem about airplanes. I asked him about the smallest Bowden cables (
https://en.wikipedia.org/wiki/Bowden_cable), and he has yet to get back to me with an answer. I will get some cables once I absolutely have the smallest available identified.
If you are not familiar, they are essentially a cable within a cable used for control linkages. They are specifically designed to transmit mechanical action/force despite bends and turns in the cabling. They are also designed NOT TO FAIL

There is an over engineering that often accompanies airplane parts because failing comes with consequences. My thought was the laces could be attached to a common bar, just like what was shown in the Nike patent at the bottom before the treads engaging with the motor. A Bowden cable could link to that common bar and be fed down into the sole. Imagine a tube in the sole that has a two ring magnets within attraction range. One is fixed at one end of the tube and the other is secured to the end of the Bowden cable. If there was a retractable catch which would retain the cable magnet, if that catch was released, the magnets would attract and the force would transmit through the cabling to the laces. The selling point here is only two magnets period. In fact, the Bowden cables could also interact with the ankle strap, but you would lose the sequence visual present in the movie. I think I have a drawing which can help the visualization of this.
I offer that idea because your design has me wondering what is the mechanism that sets the assembly in motion? My guess is you are pushing down on one of the laces and it then snaps down. I have a commitment to maintaining the visual aesthetics of the movie. So, for me at the maximum, the hands can be on the cuffs of the shoes when the laces actuate.