Not very different, since we're basically doing microkernel-y things anyway using various kinds of virtualization.
It would probably be more elegant to just run Hurd or whatever instead of for example running multiple JVMs on top of multiple Linux instances on top of a hypervisor (yes, this happens), but them's the breaks.
Well, kind of but not really. Xen isn't really bare metal in any sense of the phrase. It's just at a similar (but different) level of abstraction. In an embedded system you will need to have device drivers, respond to hardware interrupts, etc. On Xen you make hypercalls.
So eLua and these projects are similar in, say, the same way that a typical RTOS is similar to running Linux on AWS. They both have kernels, but the environments are vastly different.
It would probably be more elegant to just run Hurd or whatever instead of for example running multiple JVMs on top of multiple Linux instances on top of a hypervisor (yes, this happens), but them's the breaks.