I'd also say that you have to consider the SERP CTR and CRO aspects. By prominently displaying "Oklahoma City" in your title, you're probably scaring off customers outside of that general area, especially organic search visitors. So, the small boost you get for SEO may actually be hurting you in final sales.

I'm afraid there's no one "right" answer, and people have covered the issues pretty well:

(1) Local SEO in 2012 goes far beyond just keyword targeting, and is incredibly complex. Your TITLE tags are probably only a small part of your ability to rank locally.

(2) Any keyword you target, generally speaking, means not targeting some other keyword. If you put your city in the TITLE, you do send a signal to Google. In addition, the tag gets longer, which impacts the other keywords. That said, though, it's only one small part of the puzzle.

For traditional, organic SEO, I don't think signalling a city in the keywords automatically penalizes you for areas outside of the city. For local SEO, though, it's a different matter. If you strongly establish yourself in one area, it does imply that your less relevant for other areas. The trick is whether that's a 2% impact or 20%. Honestly, I don't think any of us could tell you in the scope of a Q&A.

It depends on how the affiliate links are set up. Let's say your affilaites using a tracking parameter (like "affiliate="). Now, you link them to a product page, such as:

www.example.com/product1.php

You could end up with a bunch of indexed pages...

www.example.com/product1.php

www.example.com/product1.php?affiliate=1

www.example.com/product1.php?affiliate=2

...etc. Those would all be seen as duplicates. Again, I'm only speaking in generalities. I don't know how your links are currently set up.

There are mixed opinions on using it on every page, but I think it's very useful on the home-page, for exactly the reasons that @donford suggests. It's easy for the home-page to get a bunch of variants indexed, including tracking parameters.

Originally, Google said that canonical wasn't proactive, but they've eased up on that. Worst case, they may just ignore it, but the All-In-One SEO approach on a blog isn't a bad bet. It's just so easy for dynamic sites to spin off duplicate URLs that it's better to be proactive.

I've never seen a penalty or devaluation due to using canonical when it's not necessary. I think Bing implied that they may ignore it if they see it too often, but I've never even seen a concrete example of that happening. It's so commonplace now that you'd hear about it if sites were being penalized.

It's less common than when you're using obvious paid links, but there is some amount of risk. It really depends on how strong and diverse your link profile is. If you're talking a relatively small percentage, it's probably ok. If the majority of your links are affiliate links, it could look shady to Google. It depends on the nature of the links, the industry, etc. too.

The other issue is that affiliate links can often create duplicate content issues, so it can make sense to consolidate them (usually, with canonical tags or 301-redirects). That's a separate issue, though.

You might find this post from Joost de Valk interesting:

http://yoast.com/affiliate-links-and-seo/

Edit: A couple more posts - it's a complex topic:

http://www.darrinward.com/blog/seo/google-penalty-nofollow-affiliate-links

http://www.wolf-howl.com/affiliate-marketing/how-to-mask-affiliate-links/

If you're targeting for Russian queries on Google.ru and your target audience is primarily entering queries with Cyrillic characters, then then Cyrillic URLs should be ok. It used to be that non-Latin character support was poor, but I think that's changed a lot over the past couple of years.

Here's a relevant Google support thread where John Mu chimes in:

http://www.google.com.ag/support/forum/p/Webmasters/thread?tid=489ece0479e0d33d&hl=en

Technically, Google can crawl/index these pages. For example, the Russian version of Wikipedia seems to be using Cyrillic URLs:

http://ru.wikipedia.org/wiki/%D0%9A%D0%BE%D0%BC%D0%BF%D1%8C%D1%8E%D1%82%D0%B5%D1%80

(unfortunately, that URL does get broken when I cut/paste)

The big question to me would be whether searchers are in the habit of using Latin characters in searches, and whether those searches draw more volume than Cyrillic. Unfortunately, we don't have any Russian speakers here on staff, so I can't comment on that one. I do speak a little Mandarin Chinese, and I've seen a mix in that market, too. Some URLs use simplified characters and some use Pinyin (the Romanized version). Technically, either should work, but there are still some legacy effects of the times when only Latin characters were supported.