"How accurate are people in estimating their mortality risk?" "How well can people estimate smoking's impact on their mortality?" (I grouped these.)
Not accurate at all. One problem with your question is that the accuracy of predicting/estimating risk is difficult to do without the actual outcome: if a smoker predicts he will get cancer, and does develop it, that's 100% accuracy. (See  and .)
Many studies show that smokers underestimate their relative risk of all smoking-related disorders (including cancer, heart disease, chronic lung infections, etc.) compared to non-smokers. Also, smokers believe they have a lower risk of developing lung cancer than the average smoker. This has been labeled by some as “unrealistic optimism”. 
Together, the accumulated data demonstrate convincingly that smokers have a very imperfect understanding of the risks of smoking and of risk statistics in general. Furthermore, regardless of what they may acknowledge about the risks faced by other smokers, they believe that their own risk is less. Given the accumulated evidence, the argument that people begin to smoke or continue to smoke with adequate knowledge of the potential risks appears indefensible. 
Smokers do react behaviorally to bad news. In one study of smokers who received annual spiral CT scans x 3, 48% of smokers quit after three abnormal scans; 28.0% quit after two abnormal scans; 24.2% quit after one abnormal scan (compared to 19.8% with no abnormal screens). A larger, later study (the Danish Lung Cancer Screening Trial [DLCST]) confirmed that quit rates were higher and relapse rate lower among subjects with initial CT findings that necessitated a repeat scan 3 months later.
Regarding other forms of "bad news", the effect was less pronounced:
Of the fifteen included studies, only two detected a significant effect of the intervention. Spirometry combined with an interpretation of the results in terms of 'lung age' had a significant effect in a single good quality trial but the evidence is not optimal. A trial of carotid plaque screening using ultrasound also detected a significant effect, but a second larger study of a similar feedback mechanism did not detect evidence of an effect. Only two pairs of studies were similar enough in terms of recruitment, setting, and intervention to allow meta-analyses; neither of these found evidence of an effect. Mixed quality evidence does not support the hypothesis that other types of biomedical risk assessment increase smoking cessation in comparison to standard treatment. There is insufficient evidence with which to evaluate the hypothesis that multiple types of assessment are more effective than single forms of assessment.
I hope this is enough to get you started.
 Smokers’ unrealistic optimism about their risk, N D Weinstein et. al.,
Tob Control 2005;14:55-59 doi:10.1136/tc.2004.008375
 Perceived risks of heart disease and cancer among cigarette smokers, Ayanian JZ, JAMA. 1999 Mar 17;281(11):1019-21.
 Relation between smoking cessation and receiving results from three annual spiral chest computed tomography scans for lung carcinoma screening, CO Townsend et. al., Cancer. 2005 May 15;103(10):2154-62.
 Effect of CT screening on smoking habits at 1-year follow-up in the Danish Lung Cancer Screening Trial (DLCST). Ashraf H et. al., Thorax. 2009 May;64(5):388-92.
 Biomedical risk assessment as an aid for smoking cessation. Bize R, et. al., Cochrane Database Syst Rev. 2012 Dec 12;12:CD004705.