There's a lot more to it than just getting a rocket up.
e.g. the biggest NK nuclear test so far was about 8 kilotons. The Davy Crocket, an American nuke that's launched from a Mortar-like device by hand has a yield iof 10-20 kilotons. So they can blow up things that are about half the yield of the WWII bombs. Sure that's still scary, but just being able to replicate designs the USA made 70 years ago doesn't imply a necessary genius at engineering, and doesn't necessarily mean they've miniturized whatever they have enough to get that payload on their 3400 km rocket.
Then, with the targeting of Guam, it's 10 times the distance as the V2s flew, and they hit within an 18 km accuracy. Guam being 550 km^2, you'd need to hit within a 23 km accuracy, so you need basically the same accuracy as the V2 but over 10 times as far. There's no evidence that they have this tech or anyone specialized in it at all.
Next, you have reentry, none of the NK long range missiles have had a payload that's survived re-entry yet. It's unlikely they have anyone who has expertise in this particular area so they're just winging it by trrial and error right now. There being less atmosphere in the way isn't a selling point. You go up and come down a lot faster meaning much more air resistance, and the need for specialized re-entry tech.
The thing is, you gotta hit Guam within a few km, then the payload needs to carry one of their nukes, and it needs to detonate at the right altititude to be sure of doing damage. All of that sounds pretty unlikely right now. They have one scientists who's good at getting missiles to launch, doesn't necessarily mean they have the technical eco-system to make everything else work, and once you have it, it's expensive to maintain that capability. America doesn't launch any ICBMs, but just maintaining the ability to do so is very expensive and requires huge amounts of manpower / brainpower.