Primordial radionuclides can be found in all environmental compartments. Since coal-fired power plants (CFPP) can be a source of additional radionuclide contamination because coal contains natural radioactive isotopes such as 238U (226Ra) and 232Th. This study investigated the impact of such possible radionuclide contamination from former heavy industrial activities, namely a former local coal-fired power plant, in urban soils and attic dust in Salgótarján, Hungary. Even today, industrial by-products, e.g., coal ash, in this city represent significant threat to its residents. A total of 36 attic dust samples (family houses, kindergartens, churches and blockhouses) were collected and 19 urban soil samples (playgrounds, kindergartens, parks and others) were selected no further than 500 m from the corresponding attic dust sampling sites. Additionally, a coal ash and a brown forest soil sample were also collected to differentiate between the anthropogenic and geogenic sources in the residential area. The sampled houses, built between 1890 and 1990, are considered to be representative sampling sites for long-term accumulations of attic dust. The mean values of the total U, Th and Cs (mg kg−1) concentrations as well as those of K (m/m %) in attic dust and urban soil samples are 2.4, 3.6, 1.7 and 0.6 and 1.1, 4.4, 1.2 and 0.3, respectively, measured using ICP-MS. The mean activity concentrations of 226Ra, 232Th, 40K and 137Cs in attic dust and urban soil samples are 43.3, 34.0, 534.4 and 88.5 and 25.1, 32.8, 386.4 and 5.6 Bq kg−1, respectively, by using a low-background iron chamber with a well-type HPGe and a n-type coaxial HPGe detector.The elemental compositions (U, Th) and activity concentrations (226Ra, 232Th) along with their abundances in coal ash from the CFPP increase in both studied media as the distance of the sampling sites from the CFPP decreases. Two outlier attic dust samples in particular show significantly high activity concentrations of 226Ra: 145 and 143, of 232Th: 83 and 94 Bq kg−1, which can be considered as a proxy of unweathered coal ash. The calculated total absorbed gamma dose rate (D) and annual effective dose (E) received from urban soils indicate that the presence of the CFPP, coal ash cone and slag dumps does not cause an increase in the level of background radiation in Salgótarján. However, the concentrations of the studied radionuclides are much higher (except for 232Th) and exhibit higher degree of variability in the samples of attic dustthan in those of urban soils. The study suggests that attic dust preserves the undisturbed ‘fingerprints’ of long-term atmospheric deposition thanks to its chemical and physical properties unlike urban soil.