Let's consider the charged non-rotating black hole since this is the simplest case. The geometry is described by the Reissner-Nordström metric, though we won't need to go into the gory details to get a basic idea of what is happening.
If you start with zero charge this is just the Schwarzschild metric. All the mass/energy is at the singularity so no matter how close you get to the centre of the black hole all the mass is still ahead of you pulling you inwards. The result is that once you pass through the horizon the $r$ coordinate becomes timelike and remains timelike as you approach and eventually crash into the singularity. No stable orbits are possible inside the horizon.
But once you charge the black hole the electric field is present both inside and outside the event horizon, and this field has an energy so it produces a gravitational force. When you're at some distance $r$ from the centre there is a part of the field behind you pulling you outwards. And the closer you get to the singularity the more of the electric field is behind you pulling you outwards.
The result is that a charged black hole has two horizons. As you pass through the outer horizon the radial coordinate becomes timelike and in this region you are doomed to fall inwards. However there is a second horizon marking the point at which the energy of the field outside balances out the mass at the singularity. As you pass through this horizon the radial coordinate becomes spacelike again and inside the second horizon you are not doomed to fall inwards. Indeed it's possible to find worldlines that travel in through the two horizons then turn round and travel back out again. For more on this see Entering a black hole, jumping into another universe---with questions.
The paper contends that inside the second horizon there are stable orbits where planets could orbit and life exist. I haven't been through the paper so I can't comment. It isn't immediately obvious that stable orbits exist inside the second horizon, after all a Schwarzschild black hole has no stable orbits for $r_s \lt r \lt 3r_s$ even though this is outside the horizon. So the fact the $r$ coordinate is spacelike doesn't guarantee stable orbits exist. However since the paper was peer reviewed I assume they have done the sums correctly.
However we should note that it is extremely unlikely a black hole would ever accumulate enough charge to move the inner horizon any great distance from the singularity, so realistically this is never going to happen for a charged black hole. With a rotating black hole it is more feasible, though I'm unsure if the supermassive black holes at galaxy cores are rotating fast enough to make it possible.