A few SAK examples from the top of my head:

Buffer http://bufferapp.com

Segment.io http://segment.io

Zapier https://zapier.com/

ITTT https://ifttt.com/

DuckDuckGo https://duckduckgo.com/

Your exactly right, with the fact that those features serve 1 purpose and 1 niche.

To clear a little confusion by what I mean on unrefined products, I can think of a prominent Swiss Army Knife...Google.

If I asked you to convey to me what is Google as if I didn't know, you'd probably say they're a search engine. While I don't mean to discredit all the products Google has, they themselves have been known to kill off their own products (Reader, Checkout, Buzz, Wave, ect) to refine a centralized purpose (ex. if you sign up for youtube, you now have a google plus).

This is the art of the niche, bringing multiple features to form one cohesive purpose.

Yes. For example, I'm working on a language learning web/mobile app, where the value proposition is that you can use this one "Swiss army knife" tool to keep track of everything, to study in a consistent way, and that works on most devices all of the time. There are lots of great tools for language learning out there, lots of flashcard apps, apps for learning to write Chinese characters, for learning pronunciation, a great site with subtitled videos etc. Learning a language takes many years though, and throughout that time you'll be using lots of tools and lots of different resources, but they're all walled gardens and isolated, and mostly work for just one or two platforms. Switching between them and keeping data synced between all of them is a pain. Also, who knows if you'll be using an iPhone forever, and whether they'll ever come around to making an android version of that app you need?

I think this is a case where the integration of all the features and data is more than the sum of its parts. The difficult part is knowing what the essential features are, and providing something useful until all of the necessary features are there.