Scientists measure their impact on the scientific community by their number of publications and how many times those publications have been cited within other publications. To an academic, publications are kind of like a form of notoriety based currency. But it's not just about how many publications you have, the quality and impact factor of the journal it's published in is important as well. Basically, someone who has 3 publications in Nature, Cell and Science (impact factor > 30) is better off than someone who has 20 publications in Animal Biology (impact factor of 0.614). Not only will an article in Nature, Cell or Science get read, and likely cited, by a wider academic audience but publications in those journals are also more likely to be picked up by the media (which could be a good thing or a bad thing...). Journals with an impact factor over 5 can still have a lot of impact, just maybe not expanding into the general public like 20+ journals would. But, in the blossoming age of open-access, its getting a lot easier for anyone, not just academics, to get their hands on scientific literature (which, again, could be a good thing or a bad thing... and could change what we deem as "impact").
Getting published is also a time consuming process. Peer-reviewed journals are considered better than non but can take months for a manuscript to get through the review process. A journal with a quicker turn-around may not have as high of an impact factor, possibly due to more lax or no review process, but could get your results out to the world faster, leading to people citing you sooner. Meaning, for the right study, the benefits from publishing in a mid-tier journal with a quicker turn-around could outweigh the benefits of publishing in a top-tier journal. So, when publishing, a scientist has to weigh the pros and cons of quantity, quality and timing.