I know you've considered it done, but since you posted it just today, I'll just fire away nevertheless.

Firstly, can we see the photographic reference you've used, so we can offer proper critiques?
I think her neck is a bit wide, but that could be just your interpretation of the light. Also on her neck, there're 2 lighter spots that seem out of place.

Another thing: your dither is making way too noisy transitions, for that much skin colours.
Maybe these things are contributing to the issue:

1. On the top circle, a colour darker than colour 1 is touching colour 2, creating cleared borders between pixels.
2. The transition between colour 1 and 2 could be less drastic. On the other hand, the contrast between colour 2 and the darker tone is quite bland. Lightening colour 2 could be the solution.
3. Try to study more dithering patterns, so you'll have enough tools to make transitions smoothly when things get rough. For instance:


Anyway, I hope your friend enjoys her gift!

Thanks Dex and Linx!
The thing is, this just is a "humanization" of a cartooney character (Tomelirolla) from a group of Brazilian animators called "Mundo Canibal", so I don't want to invest too much time on it.

But I will definitely start to show you more stuff on this topic. 

It's clear to me that in the pursuit to emulate Helm, you ended up making your own thing with a Helmish influence and consistently improve your art along the way. I don't know if it was on purpose, but as you chose this specific type of head tilt, nose length and facial features ratio, it holds a remarkable resemblance to Byzantine iconographic art. Very cool!

Wow! Super-gorgeous! I love it (mesh included).
I could definitely recognize Rose McGowan even if I didn't know the character.

Nitpickingly speaking, I noticed you have narrowed her head breadth a bit (oh, it's quite wide!), and that has given her a faint Teri Hatcher aura. As I said, just nitpick silliness, to make my comment less uninteresting.

Focusing on the dither itself: IMO, I think you're using it gratuitously.
As far as I could understand this technique (I'm also studying it), logic tells me that dithering would be useful and justifiable in three situations and both involve working with very limited palettes:

1. You may dither two colours together (50% - 50%, like you did on your 1st image) to create the illusion of a third colour. In your case, some colours were created exclusively to be used in the dither, thus defeating the purpose of expanding the palette! In the beak for instance you could've simply just used a lighter solid (non-dithered) orange as shadow, since it isn't required anywhere else as a dark orange.
2. Also, when you don't have loads of colours available, you inevitably will have to put two clashing colours together and make transitions between them look as smooth as you can. Arachne's work is definitely a good reference on this subject. Make sure you check it out!
3. You may also mix both of them! But it's hard as hell.

Check out the image bellow. It uses a Amstad CPC 16 colours palette, wich is pretty small for 160 x 200. It exemplifies all dithering features and looks pretty good.

(Image by Tocky)

Finally, it's needless to say that some artist have the ability adapt these "rules" and produce the most beautiful pieces (and that's the whole point, isn't it?).
So start studying them, but never be afraid to do lots of experimenting on your own.