I built my own version of this engine.
http://www.homemodelenginemachinist.com/index.php?topic=2106.msg17090#msg17090
I used Jan's plans as a guide but altered things to fit the materials I had to hand and made changes to facilitate experimenting with various things. Mine runs but it's finicky and frustrating.
Just a few observations based on my build of this engine...
I seriously doubt that the shape of the end of the glass tube is important. Ignore that for now.
The size of the hole in the restrictor is important. I made my restrictor to accept interchangeable bushings so I could try different size holes. Consider making up two additional restrictors - one 25% larger (in area) hole and one 25% smaller than what you're using now. Try these and see if one of them brings the engine closer to running. If it does, make two more with a similar up/down variation from the size of the one that improved things and repeat the experiment.
In mine, the distance between the restrictor and the piston at TDC affects performance.
I made the mount that supports the test tube adjustable so that this distance could be easily tweaked. You might want to experiment with that.
I can't prove it but it seems to me that this engine has very little power near BDC. To me this implies that the flywheel needs lots of inertia to carry through. You might want to try a larger flywheel.
On mine one needs to really give the flywheel a good spin to get it running. I've considered, though never tried, using a rubber wheel in an electric drill to spin it up for startup. Perhaps something else to try?
Good luck in your efforts. I hope you can get it running. You'll be very proud when you do. The engine's mind-boggling simplicity and total absence of valves will astound your mechanical friends.
Please keep us informed about your progress. I'm sure there are many forum members and lurkers who are interested in what you are doing.