The next step is an initial approach and overshoot, but if you have to make a circuit, you may as well do another eye-level pass and get as much information as you can.
Turn in at around 50-60 knots (at the 100 feet), taking particular note of escape routes, up and down draughts and turbulent areas. Maintain a constant angle, aiming directly for the point you wish to land on, controlling your speed with collective and avoiding any last-minute corrections. The idea is to keep the fuselage as level as possible, so don't move the cyclic at all, if you can help it. One reason for using the collective in this way is to minimise large control movements in the final stages, as this is a shallow approach.

Approaches
There are several schools of thought about these, but no real standard – as with many other activities involving helicopters, there is more than one “right” answer to this one. A fairly flat, disc-loaded (shallow) one will (in theory, anyway) minimise collective for the hover, and give you the most control as you keep translational lift as long as possible, but there’s very little up your sleeve at the end, and you need to be very aware of your winds, as forward speed will mask the effects right to the last minute, although it does give you a good idea of the level of your site. This assumes you remember your training and keep going forward and down, so the cyclic is ahead of the game and operating in the cleaner air in front of the machine that helps with translation. In other words, keep the rotor disc forward, so the flow of air is from front to back, especially where snow is concerned, but you shouldn’t use the shallow approach with powdered snow anyway, because you will lose sight of your landing point at the critical moment in the resulting white cloud. The other thing to bear in mind is that you are trying to land at probably the only spot available, and if you have a problem in a shallow approach, you aren't going to get there.
You could, on the other hand, use a steeper angle, particularly if you're going into a clearing surrounded by tall trees, increasing with the wind strength, but this requires large handfuls of power and attitude changes in the final stages if you don’t get ground effect, so you wouldn’t try this in an underpowered piston-engined machine that really shouldn’t be there in the first place—the engine may be able to cope with it, but can your tail rotor? (leading with the pedals will help). Anyway, since ground effect reduces your torque requirement for the hover by up to 15%, if you approach in such a way that you need no more than that amount to stop, you should find your descent stopping nicely in the right place, assuming the surface is conducive to it, and whether you have high skids or not. You also have some potential energy available for an escape.